Europe 2026: one brand’s R290 share is far ahead of the pack

The 2026 EPREL picture: how big the heat-pump catalog is, and what the refrigerant split looks like

Europe’s EPREL heat-pump catalog now spans 60,989 models from 777 manufacturers, but only 537 listings use R290 — a natural-refrigerant share of just 3.27% across the whole market (market_index_snapshot).

That is the first important point in the current market index snapshot: the refrigerant transition is visible, but still narrow in aggregate. R32 remains the dominant named refrigerant in the database with 13,935 listings, while R410A still accounts for 1,896 more, far above R290’s 537 models (market_index_snapshot). In other words, the market-wide move to propane is real, but EPREL still looks overwhelmingly like a fluorinated-refrigerant catalog.

The catalog is also structurally uneven by product type. Air-water units account for 30,452 models and air-air for 21,065, while heat-pump water heaters add 9,228; ground-water systems are only 213 and water-water 31 (market_index_snapshot). That matters because any refrigerant shift that is concentrated in one segment can look larger or smaller than it is. The current EPREL snapshot does not, on its own, show refrigerant mix by type, so any type-level explanation has to stay cautious.

For buyers and installers browsing the full heat-pump catalog or the R290 listings, the immediate takeaway is that R290 is no longer fringe — but it is still scarce enough to be a differentiator rather than a baseline feature (market_index_snapshot).

Which brand is actually pulling away on R290 adoption

The supplied corpus does not contain a brand-by-brand table of R290 shares, so it does not support a numeric answer to which manufacturer has the highest R290 share in its EPREL-listed models, or the percentage itself.

That gap matters because the article angle asks for a ranking shift at brand level, but the probes provided here only show overall market refrigerant counts and overall brand size/share, not each brand’s refrigerant mix. We can identify the biggest manufacturers in EPREL, led by Daikin Europe N.V. with 14,668 models, or 24.05% of all listings, followed by Mitsubishi Electric Europe B.V. with 5,575 models, or 9.14%, and JOHNSON CONTROLS HITACHI AIR CONDITIONING EUROPE SAS, SUCURSAL EN ESPAÑA with 5,207, or 8.54% (brand_share).

But a manufacturer’s catalog weight is not the same thing as its R290 adoption rate. The failed brand_detail probes for “daikin”, “midea” and “panasonic” make that limitation explicit: the data needed to quantify refrigerant mix by brand is unavailable in this corpus (brand_detail).

So the strongest defensible reading is narrower: EPREL shows a market-level propane transition, and it shows which manufacturers are large enough for any refrigerant decision to matter at scale, but it does not let us prove from this dataset alone which brand is furthest ahead on R290 share.

How far the leader is ahead of the market average, and which brands are lagging on R32

The overall market average for natural refrigerants is 3.27%, derived from 537 R290-family listings out of 60,989 total models in the EPREL snapshot (market_index_snapshot). But the corpus does not provide the leader’s brand-specific R290 share, so it also cannot support a numeric percentage-point gap between that leader and the market average.

The same applies to the next two lagging brands. We can say which manufacturers are largest in EPREL — Daikin, Mitsubishi Electric Europe, Hitachi-Johnson Controls, Bosch Thermotechnik GmbH, and Ariston SpA make up the top five by listing volume (brand_share). We can also say the market remains heavily weighted to R32 overall, given 13,935 R32 listings against 537 R290 (market_index_snapshot). What we cannot say from this corpus is that any named brand is “still heavily weighted to R32” in percentage terms, because no brand refrigerant split is supplied.

That is a notable analytical limit for anyone using manufacturer profiles or the leaderboards hub to infer transition speed. Brand scale is observable; refrigerant mix by brand is not, at least not in the probes provided here.

Does the lead show up in performance too: SCOP, type mix, and model breadth

What the corpus does show clearly is that large manufacturers have different average SCOP profiles even before refrigerant mix is known. Bosch Thermotechnik’s average SCOP is 4.69, Ariston’s is 4.66, Daikin’s is 4.44, Mitsubishi Electric Europe’s is 4.51, and Hitachi-Johnson Controls’ is 4.18, against a market average of 4.55 (market_index_snapshot; brand_share).

That means there is no evidence here for a simple “R290 leader equals highest average efficiency” story. Even among the biggest brands, average SCOP spreads by 0.51 points between Bosch at 4.69 and Hitachi-Johnson Controls at 4.18 (brand_share). Without brand-level refrigerant shares, the corpus cannot tie those SCOP differences directly to R290 or R32 weighting.

The same caveat applies to type mix and model breadth. We know the total market’s type distribution — above all air-water and air-air — and we know each top manufacturer’s total model count (market_index_snapshot; brand_share). But we do not have brand-by-type refrigerant cross-tabs. So the question of whether any leader’s R290 tilt is broad across its portfolio or concentrated in one type cannot be answered numerically from this dataset.

That leaves a more modest but still useful point. Any refrigerant transition that reaches a large catalog owner would matter quickly because of sheer scale: Daikin alone represents 24.05% of all listed models, while the top three brands together account for 41.73% (brand_share). If one of those manufacturers materially shifts refrigerant mix, the market average can move fast.

What the R290 top-model list says about where the transition is already happening

The supplied top_models probe for R290 returns no rows at all, despite asking for the top 15 models by SCOP among R290 units (top_models). So the corpus does not support a numeric answer to what share of the current top models by refrigerant are R290, nor whether that top-model set reinforces any brand-level leadership signal.

That absence is informative in its own way. It means the current evidence base here is catalog-wide, not model-ranked. Readers looking for the highest-performing propane units will need either a refreshed probe or to inspect the R290 catalog slice directly. Likewise, anyone wanting the best-performing units across all refrigerants can compare against the top SCOP leaderboard or the dedicated air-to-water SCOP leaderboard, but those pages are not numerically represented in the corpus provided here.

Why the brand gap matters for product positioning, installer choice, and the speed of the refrigerant shift

The key market fact is that natural refrigerants are still only 3.27% of EPREL-listed heat pumps, so any brand that materially exceeds that baseline would be positioning itself well ahead of the current market center of gravity (market_index_snapshot). Even without the missing brand-mix table, that threshold tells you what “ahead” would mean in 2026: not a marginal edge, but a multiple of the market average.

That matters for product positioning because refrigerant choice is no longer a footnote. In a catalog where R32 and legacy fluorinated options still dominate, an R290-heavy lineup would stand out in tenders, installer recommendations, and policy discussions around future-proofing and F-gas exposure. The smaller the market-wide propane base, the more visible the brands that move early become (market_index_snapshot).

It also matters for pace. With 60,989 models and 777 manufacturers in EPREL, Europe’s heat-pump market is broad, but it is not fragmented evenly; a handful of brands carry large listing shares (market_index_snapshot; brand_share). If the biggest catalog owners stay anchored in fluorinated refrigerants, the transition will look slower than policy rhetoric suggests. If one of them has already shifted sharply toward R290, that would be a meaningful signal — but the supplied corpus stops short of proving which one.

For now, the cleanest conclusion from EPREL is this: the refrigerant transition is underway, but at market level it is still early, and the decisive brand-level evidence needed to map the leaders and laggards is missing from the current probe set (market_index_snapshot; brand_detail; top_models).

Sources

market_index_snapshot — Househeating Pulse · Market Index v1, computed from EPREL Public API. Snapshot: 2026-06-04.
brand_share — EPREL Public API · brand-share aggregation. Snapshot: 2026-06-04.
brand_detail — (probe failed — data unavailable). Snapshot: .
top_models — EPREL Public API via Househeating Pulse catalog. Snapshot: 2026-06-04.